Challenging Behaviors for the Caregiver

A New Account Of Sundown Syndrome

Drake L, Drake V, Curwen J (1997) A new account of sundown syndrome. Nursing Standard. 12, 7, 37-40.


If you want to save this document to hard drive or disk remember to download the chart as well figures

In this study, the authors tested the hypothesis that residents requiring nursing care rather than residential care in a dual registered nursing and residential home would experience increased levels of confusion and agitation as the evening drew in. This article describes how the hypothesis was examined and the action that was taken as a result

Date of acceptance: October 15 1997.

Lee Drake BSc(Hons), is Researcher and Trainer; Veronica Drake RGN, is Sister and Night Services Manager; and Jackie Curwen is Care Assistant, the Sands Nursing and Residential Home, Morecambe, Lancashire.

SUNDOWN SYNDROME is widely recognised although under researched. In the US, it has been acknowledged as a medical condition in several areas of health care, including nursing homes, acute and psychiatric hospitals and the community (Laxon et al 1982). It is a syndrome of recurring confusion and increasing levels of agitation, which coincide with the onset of late afternoon and early evening. The syndrome is thought to be related to the presence of dementia, although the ability to function at a cognitive level may be retained.

It has been established that sundown syndrome and other confusional states can appear if the depleted cerebral reserve of dementia is challenged by stress or disruptive situations (Lipowski 1983, Warshaw et al 1982). Sundown syndrome resembles delirium insofar as patients with either condition may demonstrate, along with depleted cognition, other symptoms such as reduced attention, altered sleeping and waking patterns and disturbed psychomotor behaviour. All of these symptoms are more evident in the evening.

In the UK, knowledge of sundown syndrome is variable. Many professionals involved in the care of older people are unaware of the syndrome and its intricacies. This may be due to the lack of research into the condition, and a random inspection of both medical and nursing gerontology texts reveals a similar paucity of information. The result is that little is known about the syndrome's causes, its prognosis or how it may be managed or prevented.

Early studies An early report suggested that the onset of darkness was a major factor in the cause of this syndrome (Cameron 1941). In order to test this hypothesis Cameron placed known patients with the syndrome in a darkened room early in the day. In each case, symptoms of confusion appeared within one hour of imposed darkness and began to recede within an hour of exposure to light. Cameron concluded that these findings supported his theory that it was darkness, rather than fatigue, that acted as a causal factor in inducing symptoms of the syndrome.

In 1987, Evans tried to determine the prevalence of this mysterious syndrome. Evans (1987) also tried to identify and quantify the behavioural components and patterns of the syndrome, to clarify the associated medical diagnoses and to describe the existence of any other related variables. Several significant factors appeared to be shared by people with the syndrome including: greater mental impairment, increased levels of confusion and evidence of organic impairment.These people were also characterised by sharing fewer medical diagnoses, restlessness and disturbed sleep patterns.

THE STUDY OBJECTIVES
The objective of this study was to determine the existence and incidence of significantly increased levels of confusion and agitation among residents in one nursing home. The presence of any contributory factors was also sought. Evans' (1987) study provided the basis for this investigation although ours differed in two important respects. Our study was conducted over a four-day period and any subjects who were identified as suffering from the syndrome, but who displayed symptoms on only two of the days, were discounted.

Study methods The total sample for our study consisted of 52 residents of the Sands nursing and residential home. Thirty one nursing home residents consisting of nine men (29 per cent), and 22 women (71 per cent), with a mean age of 86 years, acted as the impaired group. Twenty one residents from the residential floors were used as the non-impaired comparison group, comprising seven men (33 per cent) and 14 women (67 per cent), also with a mean age of 86.

The 52 subjects had an average age of 86 years, were all white and were predominantly female (69 per cent). The majority had received education to senior school level and had resided in the home since its opening. Subjects each had an average of two co-existing medical conditions and had been prescribed an average of four medications, excluding those to take as required.

THE STUDY METHODS
Mental status of the subjects The mental status of each patient was measured using the Pfeiffer Short Portable Mental Status Questionnaire (Pfeiffer 1975). The areas covered in the questionnaire included:

The number of correct responses (up to a maximum of ten) guided the assessment of the patient's mental capacity. The mean score on the Pfeiffer Short Portable Mental Status Questionnaire was 5.8 for the 31 nursing home residents and 3.9 for the residential subjects. The combined mean for all subjects was 4.85 and equates to a moderate impairment.

The behavioural status of the subjects Those patients suspected of suffering from confusional behaviour were identified through the use of a structured observational tool known as the Sundown Syndrome Confusion Inventory (Chenitz et al 1991) (Fig. 1).

Observation of clients One or more observers placed themselves in an unobtrusive position in order to observe clients during the relevant periods. The subjects were observed for one period of ten minutes in the morning, and again for ten minutes in the late afternoon over four consecutive days. During these periods, the observer indicated the presence or absence of each of 37 psychomotor and psychosocial behaviours which were indicative of confusion and agitation. Those subjects who showed more of these behaviours in the late afternoon than in the morning consistently over the four-day period, were identified as having sundown syndrome.

The 12 subjects classified as having sundown syndrome represented 39 per cent of the nursing home group and 23 per cent of the total sample. Information about medication and medical conditions were retrieved from healthcare records and from interviews with nursing staff. Sleep patterns were observed each night for a period of one week. Any awakening was recorded along with the reason for the disturbance.

ANALYSIS OF THE DATA
From the data received, the score for the morning of each day was subtracted from the afternoon score for each day and a mean score for each day was calculated. The total sum was divided by four (the number of days) to obtain an overall mean for each day. This was performed with both the confusion inventory scores and the agitation rating scale.

This procedure was performed on each of the 12 patients identified as having sundown syndrome. The same procedure was followed for the non-impaired comparison group. The data collated from these two exercises were subjected to a parametric statistical analysis using a t-test to determine the probability of the same data arising by chance in the comparison and experimental groups.

RESULTS
The level of confusion in the group of impaired nursing home clients increased significantly with the onset of late afternoon. The probability of disturbed behaviour occurring by chance and simultaneously in both the experimental and comparison groups is very low. The results are therefore very highly significant. Figure 1 illustrates these results.

When the same statistical tests were carried out on the non-impaired residential comparison group, the findings were less significant. Figure 1 shows that a similar pattern was observed for both the agitation and confusion scales in the non-impaired comparison group. When the same statistical test was performed on the 12 patients with sundown syndrome, the results showed that these patients experienced far higher levels of confusion than the comparison group. This pattern was again observed when analysing the agitation levels of patients suffering from sundown syndrome. These patients also displayed high levels of agitation especially during the twilight and early evening hours.

DISCUSSION
The results of this study clearly identified the 12 residents of the nursing home with sundown syndrome.

Not all of the findings complied with the original project expectations. Although this present study was based on a similar investigation by Evans (1987), several important details were changed. Changes included longer observation periods, a higher behaviour threshold for accepting symptoms of sundown syndrome and a longer period of time over which the symptoms were displayed.

Evans' (1987) original study suggested that sundown syndrome was associated with the presence of dementia. Norton (1991) also claimed that this condition was generally confined to individuals with organic brain impairment, such as Alzheimer's disease and cerebrovascular disorders. However, the results of this study failed to confirm this suggestion.

Evans (1987) claimed that there were several factors relevant in the appearance of sundown syndrome in her population. Mental impairment was seen to be an important factor in its aetiology and this was confirmed in the present study. However, of the 12 patients with sundown symptoms in the present study, three were judged to be mentally intact on the results of the short mental test.

Psychosocial factors, such as whether the patient had changed rooms or whether the period of stay had been disrupted, had no contributory role in this study as most subjects had been residents for an equal amount of time. With regard to physiological factors, Evans found that all but one of the people with sundown syndrome had a cardiovascular or cerebrovascular disorder; and in addition nine of the 11 had dementia.

Depression is a common experience shared by patients with dementia (Miller 1980), and it can often be misdiagnosed as confusional behaviour, due to the fact that both share similar symptoms. This explains why depression is sometimes referred to as pseudodementia. In this study this was not the case. There was no significant condition that was shared among the sample, the only frequent condition that appeared being poor eyesight and diabetes.

Another factor highlighted in the original study was the number of medical conditions. Evans (1987) stated that people with sundown syndrome had a much lower average number of conditions (3.9 vs 5.1) than those without. However, this study found that both groups shared a similar number, 2.25. The original study claimed that sleep patterns were a contributory factor; in this study the patients also experienced sleep initiation and sleep disturbances throughout the night, and they frequently awoke for drinks and toileting.

Another contributory factor proposed by Norton (1991) focused on the role of circadian rhythms. The body clock is heavily influenced by external environmental cues. Norton (1991) raised the question that if this pattern of behaviour emerges at a specific period within the 24-hour cycle, this implies a strong connection with circadian rhythms. Fragmented sleep­wake cycles and sundown syndrome are characterised by inappropriately timed rest and activity, which may suggest that these syndromes are mediated by abnormalities of the circadian rhythm. Investigators have discovered that properly timed bright light exposure may induce phase shifts of the circadian pacemaker (Czeisler et al 1986).

It has been reported that there is an increase in sundown behaviour in nursing home patients during the winter (Bliwise et al 1989). The fact that sundown syndrome can be encouraged through restricting light together with these further findings suggest that this syndrome could be related to the perception of light combined with the functionality of circadian rhythms. Bright light treatment has improved the sleep efficiency of normal elderly subjects with sleep maintenance disturbances (Campbell and Dawson 1991) and also with some demented patients with sleep­wake disturbances (Okawa et al 1989, Hozumi et al 1990).

CONCLUSION
Although these findings do not agree with the findings of Evans (1987), they have shed new light on the mystery associated with sundown syndrome, and also raised some questions with regards to the contributory factors of this syndrome.

One final thought relating to the importance of light which has become evident throughout this study is that sundown syndrome could be related to seasonal affective disorders (SAD). People with this condition display symptoms of depression which can be misdiagnosed as dementia or states of confusion. Rosenthal (1986) treated patients with SAD with bright light therapy and found this to be highly effective. If this is the case, people with sundown syndrome could be experiencing an affective disorder which is not dependent on organic brain impairment or cerebrovascular disorders.

Further study associated with this syndrome should look at the effects of the circadian rhythm on confusion and agitation and also investigate the links between SAD and sundown syndrome.